A convenience sampling method was employed to select 657 undergraduate nursing students interning at five tertiary first-class hospitals in Henan Province, China between February 2025 and June 2025 as study participants. Inclusion criteria were: (1) full-time undergraduate nursing students; (2) having completed an internship duration of more than 1 month; and (3) providing informed consent and voluntarily participating in the survey. Ethical approval for the study was granted by the Institutional Review Board of Pingdingshan University (Approval No: 20250318), and informed consent was obtained from all participants.

A total of 687 questionnaires were distributed. After excluding 21 invalid questionnaires due to incomplete responses or obvious response patterns, 666 were retained for initial review. Upon further data checking, 9 additional questionnaires with missing key variable data were excluded, resulting in 657 valid questionnaires, yielding an effective response rate of 95.6%.

Prior to the formal survey, a pilot study was conducted with 50 nursing interns to ensure the clarity and appropriateness of the questionnaire. For the main study, the purpose, content, and significance of the research were explained to potential participants through internship group chats. Anonymity and confidentiality were explicitly guaranteed. Following the acquisition of informed consent, an online questionnaire link was generated using “Wenjuanxing” (a widely used Chinese online survey platform, equivalent to Amazon Mechanical Turk or Qualtrics in function) and distributed to the internship groups of the five target hospitals. Participants required approximately 20 to 25 min to complete the questionnaire before submission.

To test the hypothesized conceptual model, path analysis—rather than full structural equation modeling (SEM)—was conducted in AMOS. Path analysis was chosen because all variables were observed scores and no latent constructs were estimated, thereby yielding a more parsimonious and adequately powered test of the proposed mediation chain. Standardized path coefficients (β) and their significance were examined, and the significance of indirect effects was assessed with bias-corrected bootstrap confidence intervals (5,000 samples). Model fit was evaluated with χ²/df, RMSEA, GFI, AGFI, NFI, RFI, IFI, and CFI against accepted thresholds. Two-tailed p < 0.05 indicated statistical significance.

To assess the potential risk of common method bias due to the use of self-reported data, Harman’s single-factor test was performed. The unrotated factor analysis revealed that the first factor accounted for 28.7% of the total variance, which is below the critical threshold of 40%, suggesting that common method bias was not a severe threat in this study.

As shown in Table 1, the total scores for the nursing interns were as follows: Humanistic Care Ability (122.589 ± 35.693), organizational care atmosphere (94.270 ± 18.352), Communication Ability (63.242 ± 18.352), and Emotional Intelligence (91.636 ± 22.743).

As presented in Table 2, the total scores for organizational care atmosphere, communication ability, emotional intelligence, and humanistic care ability were all positively correlated with each other (p < 0.01). Furthermore, all three dimensions of organizational care atmosphere, all three dimensions of communication ability, and all four dimensions of emotional intelligence were positively correlated with both the total score and the three dimensions of humanistic care ability (p < 0.01).

The hypothesized path model (Figure 1, Table 3) demonstrated acceptable fit to the data: χ²/df = 1.176, RMSEA = 0.016, GFI = 0.984, AGFI = 0.975, NFI = 0.974, RFI = 0.965, IFI = 0.996, CFI = 0.996, meeting recommended standards.

As shown in Figure 1 and Table 4, organizational caring atmosphere had a significant direct effect on humanistic care ability (β = 0.23, p < 0.01). Communication ability and emotional intelligence also had significant direct effects on humanistic care ability (β = 0.25 and 0.30, respectively, p < 0.01). Furthermore, caring atmosphere significantly predicted both communication ability (β = 0.66, p < 0.01) and emotional intelligence (β = 0.26, p < 0.01), and communication ability significantly predicted emotional intelligence (β = 0.53, p < 0.01).

The indirect effect analysis revealed three significant mediation paths: (1) through communication ability alone (effect = 0.165, 95% CI [0.11, 0.22]), (2) through emotional intelligence alone (effect = 0.078, 95% CI [0.04, 0.12]), and (3) the sequential mediation through communication ability and then emotional intelligence (effect = 0.105, 95% CI [0.07, 0.15]). The total indirect effect was 0.348 (95% CI [0.27, 0.42]).

The results of this study indicate that the humanistic care ability score of the nursing interns was 122.589 ± 35.693, which is consistent with findings from Wang et al. (14) and Machul et al. (15). Nursing education in China has long been influenced by a predominantly technical paradigm, leading to a relative lag in humanistic education. Although recent advocacy by nursing educators, both domestically and internationally, has prompted revisions of nursing curricula—including the addition of courses such as “Humanistic Cultivation for Nurses”—the humanistic care ability of the interns in this study remained moderate. This may be attributed to the fact that the interns, recruited from various universities across five teaching hospitals, had only entered clinical practice in June. Their relatively short internship duration, unfamiliarity with the hospital environment, and lack of proficiency in nursing procedures may have limited their capacity to provide holistic, caring nursing that addresses patients’ psychological and physiological needs.

The organizational caring atmosphere score among the interns was 94.270 ± 18.352, aligning with the results reported by Yoo et al. (16). As the participants were in the early stages of their internship, positive interactions among peers, patient mentoring from clinical instructors, and tolerance from patients and their families likely contributed to a strong perception of a caring atmosphere within the hospital.

The communication ability score was 63.242 ± 18.352, consistent with the findings of Bullington et al. (17). Most of the undergraduate interns surveyed had experience participating in club activities during their university studies. These activities provided opportunities to communicate with diverse individuals, thereby enhancing their communication skills and laying a foundation for more effective clinical practice.

The emotional intelligence score was 91.636 ± 22.743, which corresponds with results from Cichoń et al. (18). Nursing interns represent a unique group undergoing the transition from academic study to clinical practice. As all participants were full-time undergraduate students with years of higher education, they possessed mature psychological development and relatively good emotional self-regulation. Furthermore, over 93% of the sample was female. While this gender distribution reflects the regional nursing-student population, caution is needed in interpreting any gender-based associations. The observed emotional intelligence level should be understood within the specific context of this sample rather than attributed to gender stereotypes.

The findings reveal a significant positive correlation between care atmosphere and humanistic care ability, indicating that a stronger care atmosphere is associated with enhanced humanistic care ability. Working in an environment characterized by care and support helps foster caring behaviors among interns. For those newly entering clinical practice, the care atmosphere is shaped by factors such as the supportive environment created by hospital administrators, the caring attitude of instructors toward patients, families, and interns themselves, and positive interactions among intern peers. However, the perception of this atmosphere is also determined by the individual’s own capacity for perception (19). Therefore, both academic educators and hospital managers should focus on enhancing the humanistic care atmosphere within clinical settings while simultaneously developing nursing students’ ability to perceive and internalize this care, thereby laying the groundwork for delivering compassionate nursing.

A significant positive correlation was also found between communication ability and humanistic care ability, suggesting that interns with stronger communication skills demonstrate greater humanistic care ability. The essence of humanistic care in nursing lies in compassionate human interaction, with effective communication being a vital component of delivering such care (20). During clinical practice, expressions of concern and caring for patients are largely conveyed through communication. Nursing schools should prioritize the development of communication skills by employing interactive teaching methods, such as scenario-based simulations and case studies, to increase student engagement and build a foundation for humanistic care ability. For students, actively participating in extracurricular clubs and engaging in communication with diverse groups can help translate theoretical knowledge into humanistic behaviors, thereby better preparing them for clinical communication.

Emotional intelligence also showed a significant positive correlation with humanistic care ability, implying that higher emotional intelligence is linked to greater humanistic care ability. Emotional intelligence is crucial for nurses working in emotionally charged environments, as it regulates their practice, interpersonal relationships, behaviors, and decision-making (21). Interns with higher emotional intelligence are better equipped to manage their own emotions in complex clinical situations, perceive the emotional states of patients and families, and implement targeted nursing interventions, ultimately enhancing patient satisfaction and fostering harmonious nurse–patient relationships.

The path analysis supports a model where the organizational caring atmosphere relates to humanistic care ability both directly and indirectly through communication ability and emotional intelligence. The sequential mediation path (Caring Atmosphere → Communication Ability → Emotional Intelligence → Humanistic Care) is theoretically plausible. In clinical settings, a supportive atmosphere may provide more opportunities for safe communication practice, which in turn could enhance interns’ awareness and management of emotions (their own and others’), ultimately contributing to humanistic care delivery. However, alternative model configurations are plausible and should be explored in future longitudinal research. The findings imply that interventions targeting the clinical environment, alongside training in communication and emotional skills, may be beneficial for fostering humanistic care.

This study has several limitations. First, the cross-sectional design precludes definitive causal inferences. The relationships discussed should be interpreted as associative. Second, the use of convenience sampling from one province in China limits the generalizability of the findings to other regions or cultural contexts. Third, despite statistical tests, the reliance on self-reported data may still be subject to biases. Fourth, the analysis used observed variable path analysis. Future research could employ latent variable SEM with multiple item parcels or indicators to better account for measurement error. Finally, alternative theoretical models were not tested; future studies should compare competing models.

Despite these limitations, the study offers theoretical and practical implications. Theoretically, it integrates organizational and individual factors within one model. Practically, it suggests that fostering a caring climate in teaching hospitals and incorporating targeted training on communication and emotional intelligence in nursing curricula could be valuable strategies. Future longitudinal or intervention studies are needed to verify these pathways and their outcomes on patient care.